This article was published in an Elsevier journal. The attached copy
is furnished to the author for non-commercial research and
education use, including for instruction at the author’s institution,
sharing with colleagues and providing to institution administration.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Available online at www.sciencedirect.com
The evolution of floral gigantism
Charles C Davis1, Peter K Endress2 and David A Baum3
Flowers exhibit tremendous variation in size (>1000-fold),
ranging from less than a millimeter to nearly a meter in
diameter. Numerous studies have established the importance
of increased floral size in species that exhibit relatively normalsized flowers, but few studies have examined the evolution of
floral size increase in species with extremely large flowers or
flower-like inflorescences (collectively termed blossoms). Our
review of these record-breakers indicates that blossom
gigantism has evolved multiple times, and suggests that the
evolutionary forces operating in these species may differ from
their ordinary-sized counterparts. Surprisingly, rather than
being associated with large-bodied pollinators, gigantism
appears to be most common in species with small-bodied
beetle or carrion-fly pollinators. Such large blossoms may be
adapted to these pollinators because they help to temporarily
trap animals, better facilitate thermal regulation, and allow for
the mimicry of large animal carcasses. Future phylogenetic
tests of these hypotheses should be conducted to determine if
the transition to such pollination systems correlates with
significant changes in the mode and tempo of blossom size
evolution.
Addresses
1
Department of Organismic and Evolutionary Biology, Harvard
University Herbaria, 22 Divinity Avenue, Cambridge, MA 02138, USA
2
Institute of Systematic Botany, University of Zurich, Zollikerstrasse 108,
8008 Zurich, Switzerland
3
Department of Botany, University of Wisconsin, 430 Lincoln Drive,
Madison, WI 53706, USA
Corresponding author: Davis, Charles C (cdavis@oeb.harvard.edu)
Current Opinion in Plant Biology 2008, 11:49–57
This review comes from a themed issue on
Growth and Development
Edited by Christian Hardtke and Keiko Torii
1369-5266/$ – see front matter
# 2007 Elsevier Ltd. All rights reserved.
DOI 10.1016/j.pbi.2007.11.003
Introduction
Flowers vary in size from less than a millimeter (Wolffia)
to almost a meter (Rafflesia), making size one of the most
variable of all floral traits. Flower size is generally
assumed to be the result of a balance between pollination
efficiency and the energetic cost of building floral structures. Several studies investigating flower size indicate
that pollinators show higher visitation rates to larger
flowers or floral displays, because they often provide
greater nectar and pollen rewards. The difference in
www.sciencedirect.com
visitation rates between smaller and larger flowers can
result in selection for increased floral size (e.g. [1!,2–6]).
Despite broad interest in size increase and its importance
in organismal evolution [7], we know relatively little
about whether flower size evolution at the extremes of
the range is governed by the same rules that act on
average-sized flowers. Here we review what is known
about floral size evolution in the world’s largest flowers,
and assess whether unique evolutionary forces are at play.
We summarize the many instances of the evolution of
floral gigantism and examine the ecological context under
which these record-breakers appear to have arisen.
Defining gigantism
The question ‘Do the same evolutionary forces act on
extremely large blossoms as act on average-sized blossoms?’ can only be answered once extreme size is defined
(for ‘blossom’ definition, see Box 1). One criterion would
be to consider blossom size relative to the pollinating
animal. This would, however, prejudge the kinds of
evolutionary forces at play, biasing our sampling at the
outset. We will focus instead on lineages possessing the
largest blossoms, exceeding "30 cm in diameter.
Although numerous plants have evolved extraordinarily
deep tubular flowers, exceeding lengths of 30 cm, our
review is focused instead on exceptional size increase in
blossom diameter as perceived by the pollinator’s
approach en-face (sensu [8]). The coevolutionary arms
race between deep flowers and their long-tongued sphingid moth [9–11] and nectar bat pollinators [12] has been
the subject of several recent investigations (including one
review), and is thus not considered here.
The evolutionary origins of gigantism
Table 1 lists 14 of the best-studied species, probably
representing at least 9 independent origins of gigantism.
It will perhaps not be surprising that several large-blossomed species are pollinated by large-bodied pollinators,
particularly bats (e.g. Adansonia) and nonflying mammals
(e.g. Adansonia, Banksia, and Protea). Additionally, some
putatively sphingid moth-pollinated flowers (e.g. Pachira)
are not just deep, but also present a large surface area to
approaching pollinators. Despite these more obvious
examples, however, what stands out in our survey of floral
gigantism is that the most striking examples involve taxa
that are pollinated by small-bodied animals such as flies
and beetles.
Why has gigantism evolved?
For those giant blossoms visited by large-bodied
animal pollinators with high energetic needs, for example
Current Opinion in Plant Biology 2008, 11:49–57
Author's personal copy
50 Growth and Development
Box 1
! Blossom: rather than restrict our discussion to flowers in the
morphological sense we also include flower-like inflorescences (as
defined in Ref. [15]) that are composed of multiple flowers with one to
many attractive floral bracts. Such inflorescences, termed pseudanthia, are functionally equivalent to a single flower, so long as the
attractive function is similar and pollinators move around the flowerlike inflorescence without flying. We would therefore assume that the
same selective forces act on multiflowered aroid inflorescences,
sunflower heads, or poinsettia cyathia as act on singly flowered roses,
tulips, or magnolias. Although the developmental constraints probably differ between pseudanthia and individual flowers (herein
referred to collectively as ‘blossoms’), the selective tradeoffs affecting
blossom size should be similar. Therefore, our concept of blossom
gigantism also includes the large, unbranched, multiflowered inflorescence of aroids such as the titan arum (Amorphophallus titanum)
of Sumatra, which superficially resembles a single giant flower and
can reach nearly 3 m in height. However, it does not include loosely
branched inflorescences, such as the talipot palm (Corypha
umbraculifera) of India, and the century plant (Agave americana) of
the North American desert southwest, which can reach 8 m. Their
inflorescences lack attractive bracts and have flowers widely spaced
enough that insect visitors would probably fly rather than crawl
between them. Additional species with large columnar inflorescences
are less easy to categorize. For example, Echium, Lobelia, Puya, and
the Hawaiian Island silverswords (Argyroxiphium) produce large
compact inflorescences, in some cases reaching several meters in
height. However, these inflorescences lack subtending bracts that
function in attraction, and floral visitation probably requires a nonnegligible energy cost for pollinators to move from one flower to the
next.
! Pseudanthium, a composite inflorescence consisting of multiple
flowers subtended by a bract or set of attractive floral bracts, which
superficially resembles a single flower.
! Cantharophily: pollination by beetles.
! Sapromyophily: pollination principally by carrion and dung-flies,
but sometimes also by carrion and dung-beetles.
night-blooming flowers that are visited by nectar/pollen
foraging bats, it seems reasonable to assume that blossom
gigantism is driven by a combination of the need to
produce and store sufficient rewards and the need to
protect reproductive organs from damage caused by large
pollinators. The cases that require more exploration are
those involving beetle pollination (cantharophily) and
carrion-fly pollination (sapromyophily; less commonly
also including pollination by carrion beetles). Rather than
artificially lump these, we will consider them in turn.
Beetle pollination: heat and traps
Beetle-pollinated blossoms are typically large in size,
often with a broad landing area and open internal space,
hence ‘chamber blossoms’ (Figure 1a,b) [13–15]. They
are also usually night-flowering, strongly scented (fecal,
fruity, musky, and nutty), pale colored (cream, pink,
white, and yellow), and offer copious food rewards (pollen, stigmatic secretions, or specialized food tissues such
as sterile male flowers). Beetle-pollinated blossoms have
Current Opinion in Plant Biology 2008, 11:49–57
evolved as many as 34 times [13] within angiosperms, and
species bearing the largest blossoms are especially common in representatives of early diverging angiosperm
lineages (e.g. Nymphaeaceae and Magnoliaceae), monocots (Araceae and Cyclanthaceae), and to a lesser extent
eudicots (Nelumbonaceae) [13,16]. In those species with
large single flowers, the basic floral form is strikingly
similar, involving many tepals surrounding a large central
chamber. Indeed for many decades the eudicot Nelumbo
(Figure 1b) was erroneously classified as a close relative of
water-lilies (Nymphaeaceae, Figure 1a) based on the
convergent similarities in their flowers [17]. One wonders
what selective forces could explain the relatively large
number of gigantic beetle-pollinated blossoms because
beetles are not especially large pollinators.
Selection tends to favor large blossoms when pollination
success increases more rapidly than the energetic tradeoffs of making a larger blossom. There could be a positive
feedback mechanism that provides enhanced pollinator
service as more beetles visit a single blossom. Increased
blossom size could maximize the number of visiting
beetles in a single flower, encouraging interactive insect
behaviors that result in increased pollination and fertilization [13,18]. For example, it may be that beetles greatly
favor blossoms that already have other beetles present
because of the increased possibility of finding a mate. The
hypothesis that larger blossoms support more beetles and
results in disproportionately increased pollination and
fertilization can be tested by determining the relationship
between reproductive success (via seed and pollen) and
the number of beetles visiting a particular sized blossom.
Another common component of beetle-pollinated blossoms is protogyny in which the female parts of the flower
mature first and pollination takes place before the male
parts release pollen. In these protogynous systems, crosspollination is facilitated by holding pollinators captive in
the blossom sometimes for over 24 hours, and then
encouraging them to move from a blossom in the male
phase to one in the female phase [16]. Such blossoms may
provide chambers (or traps) where beetles are protected
from predators and provided with food. Thus, larger
blossoms will most probably provide better protection
and more abundant food resources for trapped pollinators.
In addition to providing beetles with protection and food,
a large number of these blossoms are thermogenic and
provide heat as a reward [16] (Table 1). Thermogenesis in
floral organs not only augments scent volatilization, but
heat may also be a direct energy reward for insect visitors,
allowing visiting insects to conserve energy for feeding,
mating, and flight. Evidence for this idea comes from the
discovery that some blossoms maintain a high, nearly
constant internal temperature, despite large fluctuations
in environmental temperatures — sometimes at differentials of up to 35 8C relative to ambient [19]. Furthermore,
www.sciencedirect.com
Author's personal copy
The evolution of floral gigantism Davis, Endress and Baum 51
Figure 1
Giant blossoms: flowers and flower-like inflorescences. (a) Victoria amazonica ("40 cm in diameter); (b) Nelumbo nucifera ("30 cm in diameter);
(c) Rafflesia arnoldii ("1 m in diameter; photo # J Holden, with permission); (d) Aristolochia grandiflora (>1 m in median diameter, including perianth
appendage); (e) Stapelia gigantea ("40 cm in diameter); (f) Amorphophallus titanum (up to 3 m in height).
it is noteworthy that the temperatures found in some
thermogenic blossoms are in the same range as those
preferred by active beetles [20,21!!] (see also [22!!] for a
potentially relevant example in a bee-pollinated system).
Small blossoms with high surface-to-volume ratios are
unable to retain enough heat to raise their temperature
noticeably: studies in aroids have shown that maximum
temperature differentials between inflorescence and
ambient temperature are directly proportional to inflorescence size [23!!]. Therefore, the energetics of thermoregulation in these large chambers may also be an
www.sciencedirect.com
important reason why selection may favor gigantism in
beetle-pollinated blossoms.
It is worth mentioning here that thermogenesis in phylogenetically diverse groups is tied to a burst in cyanideresistant respiration via the alternative oxidase pathway,
for example in Araceae, Nelumbonaceae, and Nymphaeaceae [24–26]. However, the way in which thermogenesis
is triggered appears to vary. For example, while salicylic
acid is the most likely trigger to heat production in
Araceae [27], this is not the case in the distantly related
Nymphaeaceae [28].
Current Opinion in Plant Biology 2008, 11:49–57
Author's personal copy
52 Growth and Development
Table 1
The evolution of floral gigantism
Species
Size
Nearest
phylogenetic
relatives
Annonaceae
Sapranthus palanga [1,2]
Up to 15 cm long
Not known
See [3]. Sapranthus
bears large flowers
and is sister to
the small-flowered
Tridimeris (with tepals
only 4–6 mm long)
Beetles (Tenebrionidae,
Nitidulidae, and
Scarabaeidae),
homopterans (Cicadellidae),
and bees (Trigona)
Flwr
Araceae
Amorphophallus titanum [4,5]
Dracunculus vulgaris [7]
Up to 3 m in height
Up to 1 m in length
See [6]
Beetles: Diamesus
Beetles (Dermestidae,
Staphylinidae) and flies
Beetles: Erioscelis
Beetles: Cyclocephala colasi
Infl
Infl
Philodendron selloum [8–12]
Up to 30 cm in height See [13]
Philodendron solimoesense [14] 22–32 cm in height
See [13]
Aristolochiaceae
Aristolochia grandiflora [15,16]
Asclepiadaceae
Stapelia gigantea [19]
Cactaceae
Cereus peruvianus
Magnoliaceae
Magnolia macrophylla [22]
Malvaceae
Adansonia grandidieri [24,25]
Adansonia digitata [24,25]
Nelumbonaceae
Nelumbo nucifera [27,28]
Nymphaeaceae
Victoria amazonica [30,31]
Orchidaceae
Phragmipedium grande [31,34]
Proteaceae
Banksia grandis [36]
Rafflesiaceae
Rafflesia spp. [38,39]
Rafflesia kerrii [39]
Thermogenesis Pollinators
Yes
Yes
Yes
Yes
Single flower
(Flwr),
inflorescence
(Infl)
Infl
Infl
>1 m in median
diameter, including
perianth appendage
See [17,18]
Not known
Flies: Mainly Phoridae
(but also Muscidae and
Calliphoridae)
Flwr
Up to 40 cm in
diameter
See [20]
Not known
Flies: Calliphora sp.
(Calliphoridae) and Musca
domestica
Flwr
Up to 20 cm in
diameter
See [21]
Not known
Sphingid moths: Agrius
cingulatus, Manduca rustica
Flwr
Up to 42 cm in
diameter
See [23]
Not known
Various beetles,
occasional hymenopterans,
and moths
Flwr
Up to 23 cm in
diameter
Up to 27 cm in
diameter
See [26]
Not known
Lemurs
Flwr
See [26]
Not known
Fruit bats
Flwr
Up to 30 cm in
diameter
See [29]
Yes
Beetles (Chaliognathus)
and bees
Flwr
Up to 40 cm in
diameter
See [32,33]
Yes
Scarab beetles
(Cyclocephala hardyi)
Flwr
Petals more than
30 cm long
See [35]
Not known
Probably flies
Flwr
Up to 10–12 cm in
diameter and 40 cm
in length
See [37]
Not known
Birds and marsupials
Infl.
Up to 1 m in
diameter
70 cm
See [40–43]
Yes, R. tuanmudae
Not known
Female calliphorid flies
Flwr
Flies: Chrysomya
villeneuvei, C. rufifacies,
Lucilia porphyrina, and
Hypopygiopsis tumrasvini
Flwr
Current Opinion in Plant Biology 2008, 11:49–57
See [40–43]
www.sciencedirect.com
Author's personal copy
The evolution of floral gigantism Davis, Endress and Baum 53
Table 1 (Continued )
Species
Rhizanthes lowii [44,45]
Rhizanthes zippelii [45,46]
Size
Nearest
phylogenetic
relatives
Thermogenesis Pollinators
Single flower
(Flwr),
inflorescence
(Infl)
Up to 43 cm in
diameter (including
appendages)
Up to 20 cm
(including
appendages)
See [40–43]
Yes
Probably flies
Flwr
See [40–43]
Yes
Female carrion flies:
Lucilia, Chrysomya,
and Hypopygiopsis
Flwr
This table presents a list of species that exhibit extreme floral gigantism (i.e. blossoms of "30 cm or greater in diameter) as well as examples of
species that produce very large, but not enormous blossoms by our criteria. We focused especially on species where pollinator data were available.
Our list is not intended to be exhaustive, but instead provides a starting point to orient future phylogenetic and functional investigations of blossom
size evolution. In addition, we do not wish to imply that gigantism arose exclusively in these species: it may be that this trait is shared with many of
their closest large-blossomed relatives, indicating that gigantism arose earlier in the evolution of the indicated species. We have provided information
on recent phylogenetic investigations that include these species, or at least their closest putative relatives, to help pinpoint the evolutionary origins of
gigantism in future inquiries. Lastly, it is worth noting that gigantism may have evolved multiple times within some of these families. For example, this
trait is shared by several diverse genera of Araceae, suggesting that gigantism may have arisen multiple times within the family. Such a finding would
result in an overall increase in the number of times gigantism has evolved. For references see (Supplementary material).
Fly pollination: carrion mimicry
Although plant–pollinator interactions most commonly
provide benefits to both parties (mutualism), many plants
have evolved nonrewarding mechanisms [29]. Floral
mimics, which lure pollinators through deceit, act as
parasites on their pollinators. Of these, blossoms that
mimic rotting animal carcasses provide the most striking
instances of gigantism, and include the world’s largest
blossoms (Figure 1c–f): Rafflesia arnoldii ("1 m in
diameter [30]), Aristolochia grandiflora (>1 m in median
diameter, including perianth appendage [31]), Stapelia
gigantea ("40 cm in diameter [32]), and Amorphophallus
titanum (up to 3 m in height [33]). Carrion mimicry has
evolved in at least 10 lineages [15,34!!] and appears to be
commonly associated with increased blossom size
(Table 1). Although plants that mimic dung, urine, and
decaying plant material can also be large, they are often
smaller than carrion mimics.
Sapromyiophilous blossoms may exhibit striking olfactory, visual, and tactile similarity to the carcasses they
mimic [15,29,35,36]. They are most often visited by
small-bodied carrion flies and occasionally carrion beetles, which are lured into their blossoms in search of brood
sites for egg-laying and food. The blossoms are deceptive, however, in that they typically do not provide
suitable conditions for the growth and development of
insect larvae. As with beetle pollination, sapromyiophilous blossoms typically trap pollinators using contrivances such as ‘one-way hairs’, ‘slipways’, and ‘seesaw
petals’ [15]. Some species do produce nominal nectar
rewards (e.g. Rhizanthes [37] and Stapelia [32]) that apparently function in enticement, to extend pollinator visits
and/or to provide pollinators with enough energy to seek
out other blossoms of the same species. However, it has
been noted that nectar production in stapeliads may be
more important in excrement mimics than in carrion
mimics.
www.sciencedirect.com
Sapromyiophilous blossoms commonly emit strong and
fetid scents. Field and laboratory experiments have documented that carrion flies [34!!,38–43] and beetles [44]
possess highly acute olfactory systems, and are often
initially attracted over long distances by odors. The floral
volatiles that have been studied in Araceae [45,46] and
Stapelia (Apocynaceae–Asclepiadoideae) [34!!] are
remarkably diverse and extremely faithful to their models
in scent composition. Given that scent plays such an
important role in the initial attraction of these pollinators,
it is possible that large blossoms would be selectively
advantageous, providing increased scent production
[18,47!!], and a wider odor plume [48,49]. An instructive
example is Alocasia odorata (Araceae). In this species,
which attracts visitors principally through its scent
because their blossoms are visually concealed by their
vegetative leaves, the number of flies attracted to blossoms is directly correlated with the size of the scentproducing appendix [50].
Working in tandem with scent production is the fact that
many sapromyiophilous blossoms are also thermogenic
(e.g. [51–53]). Thermogenesis in floral organs in these
mimics has commonly been thought to augment scent
volatilization [36] and to make the blossom more closely
resemble a decaying animal carcass undergoing anaerobic
respiration [54]. The dragon lily (Dracunculus vulgaris), for
example, produces meter-long inflorescences consisting
of a dark purple spadix surrounded by a leathery spathe
[55,56]. When the spathe opens in the morning to reveal
its dark and festering, liver-colored surface, the spadix
begins to produce heat, and a putrid odor that is attractive
to fly pollinators. It would be interesting to know the
extent to which the pulse of heat in these blossoms
mimics the temperature and duration of heat production
in co-occurring corpses on which the mimicry is modeled.
Such coincident patterns in temperature change between
mimic and model would not be surprising because the
Current Opinion in Plant Biology 2008, 11:49–57
Author's personal copy
54 Growth and Development
decaying carcasses of bear, deer, alligator, and pig exhibit
relatively predictable and marked increases in internal
temperature several days after death, when body temperatures may rise as high as "45 8C (nearly 15 8C above
ambient) [57]. Under these circumstances, larger size
would probably be favored because larger blossoms maintain heat longer and more reliably [23!!], and perhaps
because increased heat may more accurately mimic a
decaying carcass.
Visual cues are also important in carrion mimicry and
probably work in concert with olfactory stimuli [41,43,44].
A larger blossom would be a more attractive target of
attention, particularly in the dimly lit (and dense) rain
forest understory where many of these species occur (e.g.
Amorphophallus, the largest flowered aristolochias and
Rafflesiaceae) or in the dense subtropical underbrush
(e.g. Stapelia [32]). The color and texture of carrion
mimics is also relevant: pollinators apparently cue into
their dark red to brown color, which is often accentuated
by the sharply contrasting white or yellow blotches on
their blossoms (Figure 1c–f) [41,43]. Additionally, their
tangled mat of hairs, festering pustules and darkened
orifices further add to the visual and tactile sensation
of the model. Finally, and perhaps most important, larger
carrion may be greatly preferred as brood sites by pollinators, in which case large mimics would achieve higher
visitation rates: eggs are laid at disproportionately high
rates on large carcasses, presumably because females
prefer large brood sites for oviposition [58–60]. This
may be because larval mortality is lower on larger brood
sites owing to more abundant food resources available to
developing larvae. As a result, visitation rates might be
greatly enhanced in carrion mimics by incremental
increases in blossom size — resulting in strong directional
selection for larger blossoms.
Does floral gigantism result from unique
selective forces?
We have outlined a series of hypotheses that could help to
explain why many of the world’s largest blossoms are
beetle-pollinated or carrion fly-pollinated: insect trapping,
thermoregulation, scent molecule production, and the
mimicry of large animal carcasses. As a first approach to
testing these hypotheses, it would be valuable to conduct a
rigorous statistical phylogenetic analysis to show that transitions to cantharophily or sapromyophily correlate repeatedly with dramatic increases in blossom size. Such a finding
would suggest that, indeed these pollination modes favor
large blossoms. Such studies should be complemented
with rigorous ecological and functional investigations to
evaluate whether the mode and tempo of size change in
these record-breakers is fundamentally different from that
in normal-sized blossoms.
Up until now, only one study has combined detailed
phylogenetic information with data on flower size and
Current Opinion in Plant Biology 2008, 11:49–57
sophisticated analyses of size evolution to ask whether the
tempo of size evolution could be anomalous during the
evolution of gigantism. Davis et al. [61!!] examined flower
size increase in Rafflesiaceae. They showed that these
species, whose blossoms range in diameter from "15 to
100 cm, evolved from tiny-blossomed ancestors within
the spurge family (Euphorbiaceae), and that the rate of
flower size evolution (the average change in log-size per
unit time) within Rafflesiaceae was the same as that for
closely related, small-flowered spurges. Strikingly, however, size evolution was estimated to have been 91 times
faster on the stem lineage of Rafflesiaceae than on any
other part of the phylogeny. The blossom size along this
single lineage increased at a healthy rate of "8% per
million years under the most conservative estimates,
whereas flower size showed negligible directional
increase within Rafflesiaceae or among the spurges.
Furthermore, as it is more likely that the period of
accelerated size evolution was restricted to only a small
portion of this lineage’s 54-million year duration, the rate
of increase in blossom size could have been dramatically
higher.
All species of Rafflesiaceae are believed to be sapromyiophilous [30,37,41,62,63], whereas few members of
Euphorbiaceae exhibit this pollination system. Therefore, the study by Davis et al. suggests that the switch to
sapromyophily occurred along the same branch as a
massive increase in the rate of blossom size evolution.
Although little can be inferred about evolution along the
stem lineage of Rafflesiaceae, it is easy to imagine a
coevolutionary arms race between deceptive flowers
and their deceived pollinators analogous to that driving
increases in floral depth in sphingid-pollinated flowers
[10]: larger flowers being favored in response to increases
in the threshold size for flies to visit a flower (or carcass)
and flies being selected for ever larger size thresholds
because of the negative effects of being deceived. If this
is the case, the pattern of accelerated evolution in Rafflesiaceae could be repeated in other lineages switching to
sapromyophily (e.g. Aristolochia and Amorphophallus),
something that could (and should) be evaluated in future
phylogenetic studies.
The developmental basis of gigantism
A question not answered for this, or any other system is
the nature of the developmental genetic mechanisms that
permit flower size to increase rapidly in certain lineages.
For example, in the case of Rafflesiaceae, given the
multiple origins of pseudanthial inflorescences in
Euphorbiaceae [64], we should give attention to the idea,
mentioned by Brown in his original description of R.
arnoldii [65], that its blossoms might be highly modified
inflorescences rather than single flowers. The numerous
novel features of the reproductive morphology of Rafflesiaceae (e.g. polysporangiate anthers [66] and multiseptate ovaries with many hundreds of ovules per blossom
www.sciencedirect.com
Author's personal copy
The evolution of floral gigantism Davis, Endress and Baum 55
[67]) makes one wonder if gigantism in this group did not
arise through the conglomeration of multiple floral primordia, analogous to the multiflowered inflorescences of
the sympatric titan arum. Such a change in the developmental rules of blossom development might have allowed
for unusually rapid size evolution by reducing dependence on rare mutations that increase flower size without
disrupting floral structure. Although this would be a
startling instance of developmental evolution, such a
mechanism would not be general because many other
gigantic blossoms (e.g. Aristolochia, Magnolia, and Victoria)
are unquestionably single flowers.
Conclusions and future directions
Although numerous studies have established the importance of increased floral size in species with normal-sized
blossoms, this is the first to summarize size increase in those
with extremely large blossoms. We have suggested that a
disproportionate number of examples of blossom gigantism
occur in cantharophilous or sapromyiophilous species, and
propose some explanations as to why these pollination
modes might favor the formation of massive blossoms.
Such phenomena may account for a pattern of unusually
accelerated floral size evolution as appears to be the case
with the origin of sapromyophily and gigantism in the
world’s largest flowers, Rafflesiaceae. In the future, it will
be important to conduct rigorous statistical phylogenetic
analyses to show that transitions to cantharophily or sapromyophily correlate repeatedly with dramatic increases in
blossom size. Additionally, our hypotheses for selective
forces acting on blossom size need to be tested using
manipulative and correlative field experiments to examine
the consequences of blossom size on plant fitness. Beyond
that, there is a need for further investigation on the developmental basis of blossom size increase. It is our hope that
through such research we can gain a clearer understanding
of why the natural world is blessed by the existence of such
imposing structures as the Amorphophallus inflorescence
and such spectacular flowers as Aristolochia and Rafflesia.
Acknowledgements
We thank G Adelson, W Anderson, M Blanco, P Boyce, H Cowles, A
Kocyan, M Klooster, C Neinhuis, R Raguso, and B Ruhfel for useful
comments. CCD was supported by NSF AToL EF 04-31242.
Appendix A. Supplementary data
Supplementary data associated with this article (i.e. the
references to Table 1) can be found, in the online version,
at doi:10.1016/j.pbi.2007.11.003.
References and recommended reading
Papers of particular interest, published within the annual period of
review, have been highlighted as:
! of special interest
!! of outstanding interest
1.
!
Hegland SJ, Totland O: Relationships between species’ floral
traits and pollinator visitation in a temperate grassland.
Oecologia 2005, 145:586-594.
www.sciencedirect.com
This study of plant–pollinator interactions in a temperate grassland
community established that floral visitation was strongly correlated with
visual display area: species with larger diameter flowers received more
visitors than small diameter ones. This is the first study to examine the
importance of floral traits for pollinator attraction at the interspecific level
within a community. The authors concluded that if flower size and reward
were correlated, as they are within many species (e.g. Refs. [2–6]), this
may account for the tremendous variation they documented in floral
visitation between species.
2.
Conner JK, Rush S: Effects of flower size and number on
pollinator visitation to wild radish, Raphanus raphanistrum.
Oecologia 1996, 105:509-516.
3.
Galen C: High and dry: drought stress, sex-allocation tradeoffs, and selection on flower size in the alpine wildflower
Polemonium viscosum (Polemoniaceae). Am Nat 2000,
156:72-83.
4.
Dafni A, Lehrer M, Kevan PG: Spatial flower parameters and
insect spatial vision. Biol Rev 1997, 72:239-282.
5.
Conner JK, Rush S, Jennetten P: Measurements of natural
selection on floral traits in wild radish (Raphanus
raphanistrum). I. Selection through lifetime female fitness.
Evolution 1996, 50:1127-1136.
6.
Blarer A, Keasar T, Shmida A: Possible mechanisms for the
formation of flower size preferences by foraging bumblebees.
Ethology 2002, 108:341-351.
7.
Peters RH: The Ecological Implications of Body Size. Cambridge:
Cambridge University Press; 1983.
8.
Dafni A: Advertisement, flower longevity, reward and nectar
production in Labiatae. Acta Hortic 1991, 288:340-346.
9.
Nilsson LA: The evolution of flowers with deep corolla tubes.
Nature 1988, 334:147-149.
10. Nilsson LA: Deep flowers for long tongues. Trends Ecol Evol
1998, 13:259-260.
11. Johnson SD, Peter CI, Nilsson LA, Agren J: Pollination success in
a deceptive orchid is enhanced by co-occurring rewarding
magnet plants. Ecology 2003, 84:2919-2927.
12. Muchhala N: Nectar bat stows huge tongue in its rib cage.
Nature 2006, 444:701-702.
13. Bernhardt P: Convergent evolution and adaptive radiation
of beetle-pollinated angiosperms. Plant Syst Evol 2000,
222:293-320.
14. Gottsberger G: The reproductive biology of primitive
angiosperms. Taxon 1988, 37:630-643.
15. Faegri K, van der Pijl L: The Principles of Pollination Ecology. edn 3.
Oxford: Pergamon Press; 1979.
16. Thien LB, Azuma H, Kawano S: New perspectives on the
pollination biology of basal angiosperms. Int J Plant Sci 2000,
161:S225-S235.
17. Löhne C, Borsch T, Wiersema JH: Phylogenetic analysis of
Nymphaeales using fast-evolving and noncoding chloroplast
markers. Bot J Linn Soc 2007, 154:141-163.
18. Dieringer G, Cabrera L, Lara M, Loya L, Reyes-Castillo P: Beetle
pollination and floral thermogenicity in Magnolia tamaulipana
(Magnoliaceae). Int J Plant Sci 1999, 160:64-71.
19. Seymour RS: Biophysics and physiology of temperature
regulation in thermogenic flowers. Biosci Rep 2001,
21:223-236.
20. Seymour RS, White CR, Gibernau M: Environmental
biology: heat reward for insect pollinators. Nature 2003,
426:243-244.
21. Seymour RS, Matthews PGD: The role of thermogenesis in the
!! pollination biology of the Amazon waterlily Victoria
amazonica. Ann Bot 2006, 98:1129-1135.
This study established that the Amazon waterlily, Victoria amazonica, is
pollinated by cyclocephaline beetles and that their blossoms exhibit a
bimodal heating pattern. Peaks of thermogenic activity were observed
at sunset upon arrival, continuing throughout the first night, and again
Current Opinion in Plant Biology 2008, 11:49–57
Author's personal copy
56 Growth and Development
at sunset upon departure on the second day. Chamber temperatures
were highest (between 29.3 and 34.7 8C versus ambient air of 23.5–
25.2 8C) during the first night, at which time the beetles were observed
actively competing for mates and food resources. Mean thoracic
temperatures of these beetles during this time overlapped with those
necessary for sustained activity. The authors reasonably hypothesized
that thermogenesis plays an initial role in attraction, presumably via
scent augmentation, and subsequently as an energy reward to these
visiting insects, thus allowing them to function at much lower metabolic
costs.
37. Bänziger H: Pollination of a flowering oddity: Rhizanthes
zippellii (Blume) Spach (Rafflesiaceae). Nat Hist Bull Siam Soc
1996, 44:113-142.
22. Sapir Y, Shmida A, Ne’eman G: Morning floral heat as a reward
!! to the pollinators of the Oncocyclus irises. Oecologia 2006,
147:53-59.
The Oncocyclus irises are 11 cm in diameter and possess the largest
flowers in the eastern Mediterranean. Their flowers are nearly black,
which appears to function in harvesting solar energy. The internal
floral chambers begin heating shortly after sunrise, when the temperature can reach more than 2.5 8C above ambient. Floral heat is the
only obvious reward to the night-sheltering solitary male bees that
pollinate these plants. And bees seeking refuge in these flowers
emerged earlier than bees in nonheated shelters. Bees were also
found to preferentially shelter in floral parts facing the rising sun,
further supporting the conclusion that heat is an important reward
for these pollinators.
40. Sukontason K, Sukontason KL, Piangjai S, Boonchu N,
Chaiwong T, Ngern-klun R, Sripakdee D, Vogtsberger RC,
Olson JK: Antennal sensilla of some forensically important flies
in families Calliphoridae, Sarcophagidae and Muscidae.
Micron 2004, 35:671-679.
23. Gibernau M, Barabe D, Moisson M, Trombe A: Physical
!! constraints on temperature difference in some thermogenic
aroid inflorescences. Ann Bot 2005, 96:117-125.
This paper examined thermogenesis in 18 species of Araceae. The
authors detected a significant logarithmic relationship between the
volume of the heating element in these aroids (i.e. the thermogenic part
of the spadix) and the maximum temperature difference between the
spadix and the ambient air. A larger spadix produced more heat.
24. James WO, Beevers H: The respiration of Arum spadix. A rapid
respiration, resistant to cyanide. New Phytol 1950, 49:353-374.
25. Elthon TE, Nickels RL, McIntosh L: Mitochondrial events during
development of thermogenesis in Sauromatum guttatum
(Schott). Planta 1989, 180:82-89.
26. Skubatz H, Williamson PS, Schneider EL, Meeuse BJD: Cyanideinsensitive respiration in thermogenic flowers of Victoria and
Nelumbo. J Exp Bot 1990, 41:1335-1339.
27. Raskin I, Ehmann A, Melander WR, Meeuse BJD: Salicylic acid: a
natural inducer of heat production in Arum Lilies. Science
1987, 237:1601-1602.
28. Raskin I, Skubatz H, Tang W, Meeuse BJD: Salicylic acid levels in
thermogenic and nonthermogenic plants. Ann Bot 1990,
66:369-373.
29. Dafni A: Mimicry and deception in pollination. Annu Rev Ecol
Syst 1984, 15:259-278.
30. Nais J: Rafflesia of the World. Kota Kinabalu: Sabah Parks; 2001.
31. Burgess KS, Singfield J, Melendez V, Kevan PG: Pollination
biology of Aristolochia grandiflora (Aristolochiaceae) in
Veracruz, Mexico. Ann Missouri Bot Gard 2004, 91:346-356.
38. Smith KGV: A Manual of Forensic Entomology. London, UK: British
Museum of Natural History; 1986.
39. Greenberg B, Kunich JC: Entomology and the Law. Flies for
Forensic Indicators. Cambridge, UK: Cambridge University Press;
2002.
41. Bänziger H: Stench and fragrance: unique pollination lure of
Thailand’s largest flower, Rafflesia kerrii Meijer. Nat Hist Bull
Siam Soc 1991, 39:19-52.
42. Roy BA, Raguso RA: Olfactory versus visual cues in a floral
mimicry system. Oecologia 1997, 109:414-426.
43. Wall R, Fisher P: Visual and olfactory cue interaction in
resource-location by the blowfly, Lucilia sericata. Physiol
Entomol 2001, 26:212-218.
44. Pellmyr O, Thien LB: Insect reproduction and floral fragrances:
keys to the evolution of the angiosperms? Taxon 1986, 35:76-85.
45. Kite GC, Hetterscheid WLA: Inflorescence odours of
Amorphophallus and Pseudodracontium (Araceae).
Phytochemistry 1997, 46:71-75.
46. Kite GC, Hetterscheid WLA, Lewis MJ, Boyce PC, Ollerton J,
Cocklin E, Diaz A, Simmonds MSJ: Inflorescence odours and
pollinators of Arum and Amorphophallus (Araceae). In
Reproductive Biology. Edited by Owens SJ, Rudall PJ. Royal
Botanic Gardens; 1998:295-315.
47. Valdivia CE, Niemeyer HM: Do pollinators simultaneously select
!! for inflorescence size and amount of floral scents? An
experimental assessment on Escallonia myrtoidea. Aust Ecol
2006, 31:897-903.
This study assessed how inflorescence size relates to floral scent production in Escallonia myrtoidea, a tree from central Chile with strongly
scented flowers. They used manipulation experiments to show that
inflorescence size and floral scent production were positively correlated.
48. Webster DR, Weissburg MJ: Chemosensory guidance cues in a
turbulent chemical odor plume. Limnol Oceanogr 2001,
46:1034-1047.
49. Murlis J, Elkinton JS, Carde RT: Odor plumes and how insects
use them. Annu Rev Entomol 1992, 37:505-532.
50. Miyake T, Yafuso M: Floral scents affect reproductive success
in fly-pollinated Alocasia odora (Araceae). Am J Bot 2003,
90:370-376.
51. Patiño S, Grace J, Bänziger H: Endothermy by flowers of
Rhizanthes lowii (Rafflesiaceae). Oecologia 2000, 124:149-155.
32. Meve U, Liede S: Floral biology and pollination in stapeliads —
new results and a literature review. Plant Syst Evol 1994,
192:99-116.
52. Patiño S, Aalto T, Edwards AA, Grace J: Is Rafflesia an
endothermic flower? New Phytol 2002, 154:429-437.
33. Hetterscheid W, Ittenbach S: Everything you always wanted to
know about Amorphophallus, but were afraid to stick your
nose into!!!!!. Aroideana 1996, 19:7-131.
53. Seymour RS, Gibernau M, Ito K: Thermogenesis and respiration
of inflorescences of the dead horse arum Helicodiceros
muscivorus, a pseudo-thermoregulatory aroid associated
with fly pollination. Funct Ecol 2003, 17:886-894.
34. Jürgens A, Dotterl S, Meve U: The chemical nature of fetid floral
!! odours in stapeliads (Apocynaceae–Asclepiadoideae–
Ceropegieae). New Phytol 2006, 172:452-468.
Using gas chromatography–mass spectrometry analyses of floral scent
composition from 15 stapeliad species, these authors found that scent is
much more sophisticated than previously recognized. These species
exhibit odor mimicry ranging from rotting carcasses, dung (of both
carnivores/omnivores and herbivores!), to urine.
35. Proctor M, Yeo PF, Lack A: The Natural History of Pollination.
London, UK: Harper Collins; 1996.
36. Endress PK: Diversity and Evolutionary Biology of Tropical Flowers.
Cambridge, UK: Cambridge University Press; 1994.
Current Opinion in Plant Biology 2008, 11:49–57
54. Janzen DH: Why fruits rot, seeds mold, and meat spoils. Am Nat
1977, 111:691-713.
55. Seymour RS, Schultze-Motel P: Heat-producing flowers.
Endeavour 1997, 21:125-129.
56. Seymour RS, Schultze-Motel P: Respiration, temperature
regulation and energetics of thermogenic inflorescences of
the dragon lily Dracunculus vulgaris (Araceae). Proc R Soc
Lond Ser B 1999, 266:1975-1983.
57. Watson EJ, Carlton CE: Spring succession of necrophilous
insects on wildlife carcasses in Louisiana. J Med Entomol 2003,
40:338-347.
www.sciencedirect.com
Author's personal copy
The evolution of floral gigantism Davis, Endress and Baum 57
58. Denno RI, Cothran WR: Niche relationship of a guild of
necrophagous flies. Ann Entomol Soc Am 1975, 68:
741-754.
59. Mitsui H, Kimura MT: Coexistence of drosophilid flies:
aggregation, patch size diversity and parasitism. Ecol Res
2000, 15:93-100.
60. Wall R, Green CH, French N, Morgan KL: Development of an
attractive target for the sheep blowfly Lucilia sericata. Med Vet
Entomol 1992, 6:67-74.
61. Davis CC, Latvis M, Nickrent DL, Wurdack KJ, Baum DA:
!! Floral gigantism in Rafflesiaceae. Science 2007, 315:
1812.
Rafflesiaceae include species bearing the world’s largest flowers and
have been difficult to place phylogenetically because of their unusual
morphology and anomalous patterns of molecular evolution. This study
used molecular data to confidently establish their closest relatives for
the first time since their discovery nearly 200 years ago (Ref. [65]).
Identifying the closest relatives of Rafflesiaceae played a crucial role
in determining the magnitude of blossom size increase described in the
body of the text.
www.sciencedirect.com
62. Bänziger H, Pape T: Flowers, faeces and cadavers: natural
feeding and laying habits of flesh flies in Thailand (Diptera:
Sarcophagidae, Sarcophaga spp.). J Nat Hist 2004, 38:16771694.
63. Bänziger H, Hansen B: A new taxonomic revision of a deceptive
flower, Rhizanthes Dumortier (Rafflesiaceae). Nat Hist Bull
Siam Soc 2000, 48:117-143.
64. Wurdack KJ, Hoffmann P, Chase MW: Molecular phylogenetic
analysis of uniovulate Euphorbiaceae (Euphorbiaceae sensu
stricto) using plastid rbcL and trnL-F DNA sequences. Am J
Bot 2005, 92:1397-1420.
65. Brown R: An account of a new genus of plants, named
Rafflesia. Trans Linn Soc Lond 1822, 13:201-234.
66. Endress PK, Stumpf S: Non-tetrasporangiate stamens in the
angiosperms: structure, systematic distribution, and
evolutionary aspects. Bot Jahrb Syst 1990, 112:193-240.
67. Igersheim A, Endress PK: Gynoecium diversity and systematics
of the paleoherbs. Bot J Linn Soc 1998, 127:289-370.
Current Opinion in Plant Biology 2008, 11:49–57